[MoOrB1094] Methods for estimating the costs of a nationwide HAART program in Botswana

14th International AIDS Conference

Barcelona, Spain — July 7-12, 2002

[TITLE:] Methods for estimating the costs of a nationwide HAART program in Botswana

[AUTHOR(S):] K. Ramanathan, A. Hamoudi, V. Narasimhan, R. Marlink¹, R. Otsweleng, P. Mazonde²

Int Conf AIDS. 2002 Jul 7-12;14:Abstract No. MoOrB1094

BACKGROUND: Currently, there is little data on the costs and cost-effectiveness of a fully scaled, nationwide HAART intervention in sub-Saharan Africa. As part of a larger cost-effectiveness study on HAART in Botswana, our group developed an approach to estimate the costs of health services with the introduction of HAART.

METHODS: Data regarding operating costs and capacity of Botswana's public health system were collected from clinics, hospitals, central pharmacies, laboratory facilities, tuberculosis programs and the Ministry of Health. Using these data, we developed "cost functions" to express the total cost of each health service activity as an algebraic function of the quantity of health services used. We accounted for any existing excess capacity and the scalability of new capital.

RESULTS: The cost functions indicate a high level of scale sensitivity. At 1000, 5000, and 10,000 new consultations, average clinic costs per consultation were US$ 315, US$ 64, and US$ 33 respectively; average HAART program costs per consultation were US$ 365, US$ 192, and US$ 179 respectively; and average voluntary HIV counseling and testing costs per consultation were US$ 43, US $16, and US$ 16 respectively.

CONCLUSIONS: As seen here, a high level of scale sensitivity is likely to be observed in developing country settings where the ratio of fixed capital costs to human capital and maintenance costs is much higher than in high-income countries. Cost functions improve many standard costing techniques in estimating the true resource costs associated with large scale health interventions by accounting for "scale effects," which result from factors such as large fixed start-up costs which come with the introduction of HAART, followed by low variable costs. Simply extrapolating from the costs of pilot HAART programs, therefore, may result in overestimation of the true costs of a nationwide program.

Presenting author: Kirthana Ramanathan

1Harvard AIDS Institute, Harvard School of Public Health, 651 Huntington Ave., Suite 631, Boston, MA, 02115, United States.

2Botswana Ministry of Health, Gaborone, Botswana.

020708
MoOrB1094